Forecasting Mortality in Spain: How Reliable Is Expert Opinion?

Francisco Villavicencio , University of Barcelona
Ugofilippo Basellini, Max Planck Institute for Demographic Research
Marie-Pier Bergeron-Boucher, University of Southern Denmark
Sara Solanilla Blanco, University of Barcelona
Carlo-Giovanni Camarda, Ined

Forecasting mortality is a critical component of demographic research and public policy planning, as accurate projections inform healthcare resource allocation, pension systems, and social welfare strategies. In Spain (where population ageing represents a major challenge due remarkable gains in life expectancy and persistently low fertility) official mortality forecasts rely primarily on expert opinion. While experts generally capture broad directional trends, previous research has shown that expert assessments have systematically produced conservative forecasts of life expectancy, underestimating future mortality improvements. Over the past decades, multiple statistical methods have been developed to forecast mortality based on historical data on age-specific death rates. Such methods have the advantage of extrapolating empirical data that often show long-term regularities and are more objective than expert opinion. This study critically examines the reliability of expert judgment in predicting mortality trends in Spain, comparing it to model-based approaches such as Lee-Carter and Bayesian population projections. Preliminary results show that current official mortality forecasts tend to underestimate future trends in life expectancy. Consistent with previous studies, we speculate that expert opinion remains a valuable yet imperfect tool in mortality forecasting, best used as a complement to, rather than a substitute for, statistical modelling.

See extended abstract

 Presented in Session 121. Methodological Innovations in Mortality Studies